A Thesis Submitted to the School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, in fulfilment of the requirements for the degree of Doctor of Philosophy Johannesburg, South Africa 2016. / Introduction Human immunodeficiency virus (HIV) has been associated with invasive enteric infections in HIV-infected patients, since it was first described in the 1980s. In South Africa, HIV remains an important health challenge, despite the introduction of antiretroviral therapy (ART) in 2003. In association with this, is an ongoing problem of invasive enteric infections, including those due to Shigella and Salmonella, including Salmonella enterica serovar Typhi (Salmonella Typhi). There are few South African data available as to the incidence of invasive disease due to these pathogens and how these data may contrast with the presentation and outcome in HIV-uninfected patients. The associated risk factors for mortality due to invasive enteric pathogens and whether there has been a response with ART as an intervention also needs further elucidation. Aims This work was undertaken to better describe the burden of invasive enteric infections (Shigella, nontyphoidal Salmonella and Salmonella Typhi) in association with HIV, define risk factors for mortality and establish whether the introduction of ART has impacted on disease burdens due to these pathogens. Methods Laboratory-based surveillance for enteric pathogens was initiated in 2003. Basic demographic details (age and gender) were collected on all patients where possible. In 25 hospital sites in all nine provinces, additional clinical information was collected by trained surveillance officers, including HIV status, data reflecting severity of illness, other immune suppressive conditions, antimicrobial and antiretroviral usage and outcome (survival versus death). Laboratories were requested to transport all isolates to the Centre for Enteric Diseases (CED) at the National Institute for Communicable Diseases of the National Health Laboratory Service (NHLS) in Johannesburg for further characterisation, including serotyping, antimicrobial susceptibility testing and molecular typing where relevant (whether isolates could respectively be classified as Salmonella Typhimurium ST313 and Salmonella Typhi H58). Additional cases were sought through audits of the Central Data Warehouse (CDW) of the NHLS. Annual incidence rates were calculated according to published estimates of population by age group by the Actuarial Society of South Africa for the Department of Statistics of the South African government. Analyses were specifically directed at invasive shigellosis, Salmonella meningitis, typhoid fever in South Africa and nontyphoidal salmonellosis in Gauteng Province, South Africa. Data were recorded in an Access database and analysed using chisquared test to establish differences between HIV-infected and uninfected individuals and univariate and multivariate analysis to compare risk factors for mortality. Data in the number of patients accessing ART were derived through audits of the CDW, by using the numbers of patients on whom viral loads were done annually as a proxy. Results Between 2003 and 2013, a total of 10111 invasive enteric isolates were received by CED. For patients for whom sex was recorded, 3283/6244 (52.6%) of patients presenting with invasive enteric infections were male; invasive disease was predominantly observed in children less than five years of age (1605/6131; 26.2%) and those who were aged between 25 and 54 years (3186/6131; 52.0%), with the exception of typhoid fever where the major burden was in patients aged 5 to 14 years (302/855; 35.3%). KH Keddy 81-11384 PhD iv More HIV-infected adult women were observed with invasive shigellosis (P=0.002) and with typhoid fever compared with adult men (P=0.009). Adults aged ≥ 15 years were more likely to die than children aged < 15 years (invasive shigellosis, odds ratio [OR]=3.2, 95% confidence interval [CI]=1.6 – 6.6, P=0.001; Salmonella meningitis, OR=3.7, 95% CI=1.7 – 8.1, P=0.001; typhoid fever, OR=3.7, 95% CI=1.1 – 14.9, P=0.03; invasive nontyphoidal salmonellosis, OR=2.0, 95% CI=1.6 – 2.5, P<0.001). HIV-infected patients had a significantly higher risk of mortality compared with HIVuninfected patients (invasive shigellosis, OR=4.1, 95% CI=1.5 – 11.8, P=0.008; Salmonella meningitis OR=5.3, 95% CI=1.4-20.0, P=0.013; typhoid fever, OR=11.3, 95% CI=3.0 – 42.4, P<0.001; invasive nontyphoidal salmonellosis OR=2.5, 95% CI=1.7 – 3.5, P<0.001). In all patients, severity of illness was the most significant factor contributing to mortality (invasive shigellosis, OR=22.9, 95% CI=2.7 – 194.2, P=0.004; Salmonella meningitis OR=21.6, 95% CI=3.5 – 133.3, P=0.01; typhoid fever, OR=10.8, 95% CI=2.9 – 39.5, P<0.001; invasive nontyphoidal salmonellosis OR=5.4, 95% CI=3.6 – 8.1, P<0.001). Between 2003 and 2013, ART was significantly associated with decreasing incidence rates of invasive nontyphoidal salmonellosis in adults aged 25 - 49 years (R=-0.92; P<0.001), but not in children (R=-0.50; P=0.14). Conclusion Decreasing incidence rates of invasive nontyphoidal salmonellosis and shigellosis suggest that ART is having an impact on opportunistic enteric disease in HIV. Further work is necessary however, to fully understand the associations between age, sex and invasive enteric pathogens. Specifically, this work would include typhoid fever, Shigella transmission from child to adult carer, development of invasive enteric infections in HIV-exposed children and whether the decreasing incidence rates can be sustained. Moving forward, an understanding of invasive enteric infections in the HIV-uninfected patient may assist in targeting severity of illness as a risk factor for mortality. / MT2017

HIV is a prominent infection in society and its health implications are seen in the paediatric wards daily. Despite its multi-system effect on the body, it particularly results in many respiratory infections. Effective understanding of the disease profile and management of patients with HIV relies on correct statistics and proper use of resources. Since the introduction of anti-retrovirals in 2004 in South Africa, the impact of HIV/AIDS on respiratory disease needs to be re-evaluated. The purpose of the study is to understand the disease profile of children with HIV/AIDS with regard to the presence of respiratory conditions with which they present, the need for chest physiotherapy and their health status. Of the 125 patients recruited in this study 55% were boys, average age was 20.55 months (SD= 23.64), average length of hospital stay of 2 ½ weeks (mean=18.76, SD=19.19), 80% discharged and 9.6% died. The most common respiratory conditions presented included bacterial pneumonia (66.4%), tuberculosis (48%) and pneumocystis jirovecii pneumonia (23.2%). The least common condition was lymphoid interstitial pneumonitis (4.8%). Two thirds of the children (68.8%) presented with a high burden of disease. Physiotherapy treatment was indicated for 96% of the patients mainly due to excess secretions and poor air entry. About forty percent (40.8%) of children were taking anti-retrovirals with an average length of use of 9.81 months (SD=11.61). Three out of four (75%) mothers were not involved in a PMTCT program. The analysis of immune status revealed a mean CD4 percentage 17.33% (SD=10.96), CD4 absolute 631.36 cell/mm3 (SD=610.36) and viral load 2.6 million copies /ml (SD=9.08 million copies/ml). A higher burden of disease was related to the use of anti-retrovirals, a lower immunity, female patients, longer length of hospital stay and incidences of mortality occuring at later periods of hospital stay. Results of this study highlight the characteristics of respiratory disease burden among children with HIV in a South African setting in a post HAART era.

Ph.D., Faculty of health Sciences, University of the Witwatersrand, 2011 / The rate of HIV infection in South Africa remains high despite the continued efforts to prevent its transmission. Conversely, the rate of AIDS related mortality has been on the decline since the country introduced its comprehensive care and treatment plan. Among other things, the comprehensive plan seeks to coordinate activities to prevent further infection and facilitate health and social support services for the infected and affected. Despite this, South Africa has an increasing number of people living with HIV (PLWH) and is home to the largest population of HIV infected people in the world. Consequently, there is a growing shift of focus of HIV/AIDS response programmes from issues of survival and death prevention to the quality of survival of PLWH. This point accentuates the need for patient evaluation of their wellbeing and quality of life in the context of the country’s HIV/AIDS response services. There are HIV-specific quality of life assessment tools like the WHOQOL-HIV, but these tools are subject to cultural variations and have not been validated across South Africa’s multicultural society. The available tools are often too long and cannot be used in routine care and support of PLWH. In the absence of validated and efficient quality of life assessment tools, there is over reliance on biomedical markers of HIV/AIDS like the CD4 cell count and viral load. These biomedical indicators do not provide a complete view of the impact of the disease given the multidimensional consequences of HIV and AIDS. With the above background, this study sought to firstly explore the health related quality of life (HRQOL) and the experiences of PLWH in seeking care and support services, and secondly, to validate the ability of the WHOQOL-HIV instrument to measure the health related quality of life of PLWH among the three cultural groups in Limpopo Province over a specified time period. The study was divided into two, with the first and second studies using qualitative and quantitative research approaches respectively. Both studies divided their participants into three groups, with each group representing each of the three main ethnic groups in Limpopo Province (Pedi, Tsonga and Venda). The qualitative study used focus group discussions (FGDs) iii to solicit information and grounded theory to guide its participant selection, number of FGDs conducted and the interpretation of its findings. The second study was a longitudinal follow up of participants in the three groups from the point of treatment initiation over 12 months. In the second study, there were three points of data collection (baseline, six and twelve months). The participants were conveniently selected and interviewed repeatedly with the WHOQOL-HIV and MOS-HIV instruments. The study findings were mostly presented in tables and t-tests and ANOVA were used to compare quality of life scores in different groupings while correlation and limits of agreements were used to establish instrument validity. Item-total correlation coefficient and alpha if item deleted, was used to explore the possibility of shortening the WHOQOL-HIV instrument in an attempt to suggest a shorter and more user friendly version of the instrument. The key findings of the qualitative study are that the quality of life of PLWH in Limpopo is determined by three groups of factors (physical, mental and external). The frequency and severity of these factors are determined by the participants’ duration on treatment, disclosure of their HIV status and socioeconomic status. The physical factors were mostly constituted by biological symptoms of the disease. The most commonly cited symptoms were diarrhoea, pain and fatigue. The study also noted marked differences in the occurrence of the symptoms through the trajectory of the disease revealing four main stages: pre-treatment; treatment initiation; early treatment; and treatment maintenance stages. The study also noted that the impact of the determinants of the quality of life on an individual is influenced by intervening factors which can be altered by a set of modifying factors. In all, the quality of life determinants identified by this study replicates those that constitute the WHOQOL-HIV instrument. This finding hence obviates the need for the modification of existing quality of life instruments to suit the three South African ethnic groups investigated by this study. The quantitative study showed high reliability of the WHOQOL-HIV among the three ethnic groups with alpha ranging from 0.79 to 0.94 in the six domains of the instrument. The study iv also showed that the quality of life varied by socio-demographic characteristics such as ethnic group, sex, marital status, number of children, employment status and membership of HIV/AIDS support groups. The observed difference reflects poor financial capacity and activity tolerance across the various categories of the participants and at different times. While significant changes in the quality of life was noted between the baseline data and the six and twelve months data respectively, little or no improvements were seen between the six and twelve months data. The participants were actually more likely to have a better quality of life at six months when compared to their quality of life at twelve months. In the same manner, the median CD4 cell count and viral load were very similar at six and twelve months but differed significantly with the baseline reports. Over time and across cultural groups, the participants reported lower quality of life in the level of dependence domain and financial support subscale. There were little or no associations between the biomedical markers and HRQOL indicators. In addition, the quality of life tended to increase with increase in CD4 cell count. The validation of the WHOQOL-HIV using the MOS-HIV through a correlation of similar domains and their limits of agreement largely suggests that the WHOQOL-HIV is valid but both instruments are not exact replicas in their measurements. The multidimensional nature hypothesized by the original WHOQOL-HIV instrument study was also demonstrated by the factor analysis component matrix. Attempts to shorten the WHOQOL-HIV proved to be impossible as the items with highest item-total correlation varied with the participants’ duration on treatment. In conclusion, this study suggested the needs and factors that contribute to the quality of life of PLWH in Limpopo and how those factors can be harnessed through a set of modifying factors. With this, an individual’s quality of life is determined by the delicate balance between intervening and modifying factors. The linkages between the observed determinants of quality of life suggest a vicious circle where one determinant may exacerbate the effect of another determinant. The study also showed that the WHOQOL-HIV instrument is valid and reliable in measuring the quality of life of PLWH in the province. The observed poor to no associations between the health related quality of life indicators and the biomedical makers show that they cannot be direct proxies of each other. Finally, the study discourages any further shortening of v the WHOQOL-HIV instrument on the basis that HIV-infected people are not a homogenous group as their bio-psychosocial needs vary with time and their position along the trajectory of the disease.

Thesis M. Med.(Family Medicine))--University of Limpopo (Medunsa Campus), 2010. / Background The use of HAART in HIV/AIDS patients has been recognised worldwide to improve the quality of life and survival prospects. Neverthess, factors such as WHO clinical stage III-IV, CD4< 200, VL> 100,000, anaemia, blood transfusion, malnutrition, male gender, intravenous drug use, drug toxicity, HAART experienced by patients, hospitalization, older age and depression have been reported to be associated with negative outcomes whereas, in contrast, white ethnicity, adherence > 90%, antiretroviral naïve subjects, longer period of viral suppression, younger age, and female gender have been reported to be associated with positive clinical outcomes. Methods The researcher conducted a descriptive retrospective study of 78 systematically selected patients who initiated HAART during the period of 5 June 2007 to 5 December 2008. Data regarding demographics, nutritional status, patients’ opportunistic infections, patients’ use of ARV drugs and HAART regimens, side effects and adverse events, baseline and follow up measurements of CD4 cell count, VL, ALT and Hb were collected at initiation, 6 and 12 months of HAART and analysed, utilizing descriptive statistics. Results Of the 78 patients recruited for the study, 60 (77%) were females and 18 (23%) males, 77 (98.8%) black and 1 (1.2%) coloured. The majority of patients belonged to the two age-groups 26-35 years (35.9%), and 36-45 (37.2%). The majority of patients [73/78 (93.4%)] were unemployed and residents of Vryburg town. Nutritionally, 17/78 (21.8%) patients were underweight. Clinically, 79.4% were classified as WHO clinical stage III - IV. The mean weight improved in both sex at 6 and 12 months of HAART respectively, from 57.5kg (SD 8.0) to 63.0kg (SD 13.0) and 65.2kg (SD 4.5) for males. 12 Conclusion The administration of HAART to patients attending ARV clinic at Vryburg District Hospital was followed by better clinical outcomes in terms of weight gain, correction of anaemia, increase in CD4 and achievement of virological suppression. Female gender, VL > 100,000 copies/ml, Younger age (< 46 years) and good adherence were found to have positive influence on clinical outcomes.

Thesis (M. Med.(Family Medicine))--University of Limpopo (Medunsa Campus), 2010. / A study on Factors associated with non-disclosure of HIV positive status to sexual partners by adult patients attending the VCT clinic at Nhlangano health centre, Swaziland. Aim: To describe the factors associated with non-disclosure of known HIV sero-positive status to sexual partners by adult patients attending the VCT clinic at Nhlangano health center, Swaziland. Design:- Cross-sectional study using questionnaire administered by a trained research assistant. Setting:- Nhlangano health center VCT clinic, Nhlangano town in the Shiselweni region, Southern Swaziland. Study population:- All adult patients above the age of 18 years who had undergone HIV testing, who knew their positive HIV status and had follow up visits at the VCT clinic of Nhlangano health centre, from November 2005 (when the centre started rendering VCT service) till the beginning of the data collection, in September, 2008. Results:-The vast majority (89.1%) disclosed their positive HIV status to their sexual partners and 94.6% believed that letting their sexual partner/s know about their HIV status was very important. In terms of knowing the HIV status of their partners, 55.4% knew the HIV status of all of their sexual partners and 44.6% knew only the status of the regular partner/s. With regard to condom use, 96.7% believed that using condoms helps them to prevent transmission of HIV and 91.3% expected that letting their partners know about their HIV status would help them use condom IV more frequently. Only 53.3% said they would insist on condom use even if their partner is not willing to use. Conclusion:-  The rate of positive HIV status disclosure to sexual partner found in this study compared to many studies done in other settings is considerably high. This is encouraging especially considering the existing very high prevalence of HIV infection in the country. Despite this though, knowing partner's HIV status was relatively lower. Therefore, people are more likely to share their HIV status with a partner than insist that the partner does the same.  Even though the study was done only amongst patients attending VCT, it is important to note that the majority of the patients had positive attitudes about HIV status disclosure to a partner, and believed in the importance of letting their sexual partner/s know about their HIV status. Patients understood the unethical nature of engaging into sexual intercourse without disclosing their positive HIV status to their partner. It is possible to conclude that factors which contributed to these positive results should be implemented at a larger scale, namely creating awareness, health education, good counseling and follow up of treatment.  Awareness of the importance of condom use in preventing HIV transmission (including the fact that disclosure of HIV status to a partner enhances its better use) was impressively high amongst almost all participants. But it is worrisome that only half of the participants said they would insist on condom use irrespective of their partners’ willingness to use it or not. V  The variables that were found to be independently associated with disclosure to a partner comparing those who disclosed with those who did not were gender, age, marital status, education, number of sexual partners, and stage of the HIV condition.

Thesis (MPH) -- University of Limpopo, 2011. / Introduction HIV/AIDS is major cause of child mortality and an increase in the number of sick children presenting to health services worldwide (UNICEF 2008). A significant number of children live with HIV/AIDS in South Africa. Research indicates that in poor resourced countries, there is an increase in the prevalence of hospital admissions and re-admissions among HIV infected children as compared to developed countries. Research data on hospital admissions, treatment and care of HIV positive children South Africa is limited. Objectives This study was therefore initiated to determine the demographic and clinical causes of HIV positive children admitted and readmitted at the paediatric ward of Dr George Mukhari Hospital (DGMH), South Africa in the year 2003. Methods This was mainly a descriptive quantitative study using medical records of HIV infected children admitted and readmitted in the paediatric ward of DGMH from 1st January to 31st December 2003. A full census of all the records of children admitted in the two paediatric wards was carried out. Descriptive and inferential statistics were used to analyze data. Results The study comprised 74 children, 28 (37.8%) female and 48 (62.2%) males. The average mean of initial admission length of hospital stay was 12.3 days and (SD = 12.1) days. The different diagnoses were classified in accordance with World Health Organization (WHO) Clinical Staging of HIV disease for infants and children with established HIV infection. Out of a total of 581 initial admissions, 74 (12.7%) children were readmitted. The mean interval days between the discharge date and readmission date was 9.8 days (SD = 7.0 days) and 94.6% of the readmissions occurred within the first two weeks of discharge date. Second readmission decreased by 75.3% as only 18 patients were readmitted. A further 95.9% decrease in the third readmission was noted with only 3 patients getting readmitted. iv The commonest causes of admission with HIV were broncho-pneumonia, gastro-enteritis, vomitting, oral thrush, immunosuppression with symptoms like fever,cough, respiratory distress. Causes of readmissions were broncho-pneumonia, oral thrush, diarrhoea, vomitting, immunosuppression, pulmonary tuberculosis, wasting and failure to thrive, dehydration associated with symptoms like fever, cough, respiratory distress and upper respiratory distress. Conclusion The rate of readmission was (12.7%) and majority of the readmitted children were in the 0- 2-year age group. The study results show a high prevalence of diseases of the respiratory system with a high frequency of broncho-pneumonia and a high prevalence of diseases of the digestive system with a high frequency of gastro-enteritis. The average mean of the initial hospital admission stay was 12.3 days, which was significantly higher than other studies previously conducted. The probable reason for a long hospital stay could be the high prevalence of co-infections among the children admitted.

Thesis (MPH) -- University of Limpopo, 2011. / Introduction With the increased availability of the life-saving ARVs in most Sub-Saharan Africa more HIV -infected children are surviving into their adolescent years and beyond hence giving rise to the question of whether the caregiver should disclose or not disclose the child's HIV diagnosis to child. Little is known of the reasons and experiences that motive or hinder caregivers from disclosing the HIV diagnosis to the child. Study Aim and objectives This was aimed at identifying caregivers' reasons for HIV diagnosis disclosure and non-disclosure to HIV-infected children under their care. The study also explores their experience with process of HIV diagnosis disclosure and non-disclosure to the child. Study methodology Using qualitative descriptive research approach, twenty (20) caregivers of HIV¬infected children aged between 6 - 16 years receiving ART at Thamaga Primary Hospital IDee with unknown HIV diagnosis disclo~ure status were ~.ubjected to audio-taped in-depth interviews for data capturing. Thematic content analysis was used for data analysis using, Nvivo8 software and 16 themes with their sub categories were identified. Findings Both caregivers of disclosed and non-disclosed HIV-infected children perceived disclosure as a good thing to do with majority of the caregivers (60%) having disclosed. Reasons for telling the children their HIV diagnosis were that the child had the right to know his/her status; caregiver tired of keeping child HIV diagnosis a secret; the caregiver's believe that disclosure will improve the child's ART adherence and finally some caregivers felt the child had reached the right age or maturity for disclosure. Non-disclosing caregivers felt that health care workers should assist them in doing disclosure and identified the reasons for non-disclosure as the child being too young and not asking questions about their illness; fear that disclosure might hurt the child psychologically; fear that the child might not keep their HIV diagnosis a secret leading to discrimination in the community while some caregivers lack of knowledge on how to disclose. Non-disclosing caregivers managed disclosure by not telling the child the truth about their diagnosis and using threats to coerce them to take their ARV drugs. Conclusions and recommendations The decision to disclosure or not to disclose the HIV diagnosis to a child by a caregiver is influenced by a number of reasons and their experiences. Caregivers of HIV-infected children need to be assisted by a health care provider when disclosing to the child and further assessment should be made in making disclosure part of the holistic management of an HIV -infected child. v

Diss. (sammanfattning) Stockholm : Karol. inst. / Härtill 7 uppsatser.

Diss. (sammanfattning) Stockholm : Karol. inst. / Härtill 6 uppsatser.

Diss. (sammanfattning) Stockholm : Karol. inst. / Härtill 7 uppsatser.